Which voice cloning API works directly in a TTS request?
Condition a line from reference audio alone or pair the recording with its exact transcript. You do not need to create a saved voice first.
Which voice conditioning method should I use?
Reference audio
Supply a clean, authorized recording to clone its speaker identity for one request.
Saved reference voice
Save an approved clone when the same speaker identity will be reused across requests.
Reference plus transcript
Use a clean recording and its exact spoken transcript for reference-led voice conditioning.
How do I condition or clone a voice?
Choose the reference path
Use reference audio alone or combine it with the exact reference transcript.
Keep the transcript exact
Reference transcript must match the words spoken in the supplied recording.
Generate with permission
Use only recordings and identities you are authorized to use, and preserve consent records for saved voices.
What else should I know?
Can I clone without creating a saved voice?
Yes. Send authorized reference audio directly in the speech request.
Can I add a transcript to reference audio?
Yes. Send the exact words spoken in the reference recording as reference transcript.
Can reference transcript be sent alone?
No. Reference transcript is only used together with reference audio.